Pressure to increase productivity and deliver greater business value while limiting headcount has many organizations considering data center automation. Fortunately, there are many tools available to get it done. Here are nine to know if you’re thinking about automating all or part of your data operation.
Here’s some breaking news for you: Human beings have limits. It’s hard for most of us to accept, especially when we’re being pushed to increase productivity and business impact while we limit or reduce headcount in the data center. Fortunately, there’s a way to amplify the impact of staff with data center automation.
Data center automation is one of those terms that is like the elephant in the parable of the blind men. Managers tend to focus on the aspect that is closest to their specific need and define the whole topic from that point of view. While that’s very understandable it’s also unfortunate because it means that they can be blinded to the full potential of automation.
Data center automation can be approached from many different directions, from a simple need to get a handle on virtual server creation to full DevOps to the ultimate in “lights out” data center management. While it’s possible to build from one stage to another, it’s also true that automating the entire data center operation requires a different operational and software framework than automating a single task.
All of that is, of course, separate from the issue of precisely what kind of data center we’re talking about. A data center that houses a mainframe is a different creature than one that is filled with Intel-based servers. And once you start homing significant parts of your “data center” in the cloud, well, now we’re talking something else entirely. The fact that each of these is a legitimate operational definition of “data center” makes any discussion of automation tools one that’s going to require lots of term definition and more than a little disagreement.
OpenStack
If you have been thinking about taking part of your infrastructure to the cloud it’s likely that you’ve heard of OpenStack even if you’re still trying to work out precisely what it is. According to the supporting organization’s website, “OpenStack software controls large pools of compute, storage, and networking resources throughout a datacenter, managed through a dashboard or via the OpenStack API.” That’s nice, but there’s still room for learning about the software.
OpenStack is an operating system that helps build a cloud infrastructure or (and here’s where it gets interesting for data centers) manage local resources as though they were a cloud. That means automating the building, tear down, and management of virtual servers and other virtualized infrastructure. Since OpenStack is open source, there are several distributions, many implementations, and lots of consultants and integrators just itching to help you out. OpenStack is still being developed and enhanced but that hasn’t kept many companies from embracing it as the way forward for their cloud and data center automation framework.
Puppet
Puppet is another tool that tends to come up often in conversations around DevOps. Puppet is a framework and language that system operations professionals can use to define operations like software deployment so that they can be automated. Puppet (the language) creates the definitions and workflow that are implemented by Puppet (the framework).
Two of the significant benefits that Puppet brings are a common language and compatibility across a broad swath of devices. Many IT departments are using Puppet to automate complex processes involving many different pieces of hardware and software packages. That has made Puppet one of the languages that operations professionals are learning to improve their job opportunities and that IT departments are seeking.
CloudStack
What if a company decides an open source project doesn’t deliver the functionality the business needs? What happens if an open source project can’t satisfy the needs of customers? In the case of Citrix and OpenStack, the company bought technology and released it through the Apache Incubator program. Now, CloudStack is competing with OpenStack as an open source framework for creating and managing a cloud infrastructure.
In some ways CloudStack is the superior technology. It supports a broader variety of hypervisors and network models than OpenStack, it deploys more simply, and it is highly scalable. On the other hand, since it’s newer it has a smaller community. The important thing, though, is that it gives IT managers two open source options for automating and managing data center processes.
Microsoft System Center
Microsoft Management Console started out as a way to manage Windows NT servers but the transition to Microsoft System Center brought expansion outside the world of Microsoft. With System Center 2016, Microsoft has provided a management and automation system that incorporates Linux and Microsoft servers, cloud and on-premises infrastructure, and a broad variety of compute, storage, networking, and security components.
Microsoft System Center is usable if you’re looking for a way to automate operations on a server. It’s also a candidate for the center of a hyper-converged architecture. It won’t be anyone’s choice for mainframe automation, but short of that, Microsoft has provided a management and automation framework that can be considered for most implementations.
OpenNebula
Data center automation tools are complicated because they job they have to do is complicated. That’s why it’s so unusual to find a tool that touts simplicity as one of it’s chief virtues. Simplicity is what OpenNebula is all about and for teams looking to get into automation and virtualization, that virtue may be enough to take OpenNebula to the top of the list.
OpenNebula doesn’t have the breadth of platforms CloudStack provides, but it incorporates a number of features that will not require add-ins or integration — add-ins like accounting charge-backs and dynamic creation of full virtual data centers. If you need a framework that an internal staff can attack with reasonable effort, then OpenNebula has a lot to offer.
Eucalyptus
For many organizations, today’s data center extends to the cloud in the form of Amazon Web Services (AWS). If you’re looking at automating data center operations across on-premises infrastructure and AWS, then Eucalyptus could be a tool worth exploring.
The official name of this open source tool is now HPE Helion Eucalyptus since its 2014 purchase. Born in an open grid project at Rice University, Eucalyptus treats everything it manages as a cloud instance, whether that cloud is a private, on-premises cloud, or hosted in AWS. With HPE’s purchase, Eucalyptus received more support so that it continues to bring many different virtualized infrastructure pieces in the local data center together with their AWS analogs. If your local hardware is from HPE, this is a natural tool to consider, but even if your hardware has a different badge, Eucalyptus is a solid option for AWS customers.
Chef
Chef is a suite of products that is open source and commercial — organizations can begin their work with Chef in the open source components and either grow capabilities through integration or shift to the commercial version if the requirements demand.
Chef is written in Ruby and provides a framework in which users can write “recipes” in a Ruby-like language. Those recipes can implement processes that span an entire infrastructure or focus on a single component. The three components of Chef — Chef, Inspect, and Habitat — can be used alone or together for a complete DevOps framework. If you’re currently an agile shop and want to move toward DevOps, Chef is a candidate to get you there
Ansible Tower
Ansible Tower is Red Hat’s automation platform for Red Hat Linux and beyond.
Ansible Tower is designed to be a software framework that supports disciplines ranging from agile development to DevOps to continuous delivery. If you run a Linux shop that needs to automate the data center, then Ansible Tower is the sort of package that should be on your short list of software candidates.
Git
Much of data center automation revolves around software and for many in the agile and open source communities, software revolves around Git. Git is an open source code repository and version control system that is a key ingredient in many organizations’ agile and DevOps workflows.
Git has the advantage of being widely known in the development community. There are relatively few developers who don’t have at least some experience with Git through Github. With a huge pool of platform expertise, a huge open source community, and software that is relatively easy to install and configure, Git can be a piece of your automation environment, no matter which other pieces of software you include in the infrastructure.
So there they are. Nine tools that you should know about if you’re looking to automate your data center or any significant piece of your data infrastructure. Are you using any of these? Have you tried one and decided not to use it? Let me know what you think about these tools and the others that can help when it’s time to turn automation loose on your data center.